About Reading Companion
Reading Companion answers practical questions about reading books in a language you are learning: which book fits your level, how long it will take, whether to start with graded readers, and what to do with unknown words.
Who writes it
The site is written by the team that makes Readlang Mobile, an iPhone and iPad app for reading books and articles in a foreign language: you tap a word and its translation appears in the line you are reading. We build a tool for this kind of reading, so we are not neutral about the topic. We mention our app only where it is relevant, next to other tools that do similar things, and we say it is ours each time.
Readlang Mobile is an independent app. It is not affiliated with Readlang (readlang.com), the web reader and browser extension, which is a different product from a different developer.
Questions, corrections and disagreements: support@readlang-mobile.com. If you find a wrong number, tell us where you saw it and we will fix the page and change its “Updated” date.
Where the book data comes from
The table of books lists the 117 public-domain books in the Readlang Mobile library as of September 2026: 98 editions from Project Gutenberg and 19 from Standard Ebooks, in English, Spanish, French, German and Italian. Each title links to the exact edition we measured.
How the levels were set
When a book was added to the library, a member of our team gave it a CEFR level by hand, from A1 to C2. The level is a judgement about the language of the book as a whole: vocabulary, sentence length, how old the language is, and how much the reader has to guess. It is not the result of a formula or a test with readers, and two careful people could place the same book a level apart. That is why every page that uses our levels also explains how to check a book yourself.
How the words were counted
We counted the words in each EPUB file with a script, in September 2026:
- Only the text of the book is counted, in reading order. Title pages, tables of contents, the Project Gutenberg licence and the Standard Ebooks colophon are left out. Editors’ notes, prefaces and glossaries that are part of the edition are counted.
- A word is an unbroken run of letters in any alphabet, so accented letters (é, ñ, ü, ß) are part of the word. An apostrophe inside a word does not split it (“don’t”, “aujourd’hui”); a hyphen does (“well-known” is two words).
- “Different words” are distinct word forms, ignoring capital letters. “Walk”, “walks” and “walked” are three forms. In French and Italian, an elided article or preposition is removed first, so “l’homme” counts as “homme”. Names of people and places are included.
These rules give numbers close to what a word processor would show, but not identical, and other tools will count a little differently. Different-word counts depend strongly on the length of a book, so they are useful for comparing books of similar size, not as a difficulty score.
How reading time is estimated
Reading time is the word count divided by a reading speed. The table shows a range between 200 and 100 words per minute. Why those two speeds, and how to measure your own, is explained on How long does it take to read a book in a foreign language.
How we use research
Every number that comes from a study is linked to the original publication, with the authors and the year, next to the sentence that uses it. Before a page is published, each of these numbers is checked against the original paper or, where the paper is paywalled, against the publisher’s abstract. Where research is thin or disputed, we say so instead of rounding it into a rule.
This address before us
From 2006 to about 2015, readingcompanion.org was home to Reading Companion, a free web-based literacy program from IBM that used speech recognition to help children and adults learn to read. After that the address passed through other owners and for a time carried unrelated content. We have no connection to IBM, to that program or to the later owners. We registered the expired domain in September 2026 because our topic, reading, is close to the original one, and older pages elsewhere on the web still link to this address. Links to pages of the old program now lead to this site’s home page or to this page.
Privacy
We count visits with Umami, an open-source counter that does not use cookies and does not follow you to other sites. There are no ads on this site and no third-party scripts.
Reusing our material
You may quote short passages with a link to the page. The book table is available as a CSV file under the CC BY 4.0 licence.